1. Technical Field
The invention relates to an image processing device, a robot, a robot system, and a marker.
2. Related Art
There has been research and development on a robot that detects a marker or a target object from an image and performs work according to a detected marker or a detected target object.
Relating to this, a two-dimensional code including a marker configured with an area of a first color and an area of a second color adjacent to each other and a data matrix formed from binary data cells arranged in a matrix form is known (see JP-A-2013-84031).
An image detection method for binarizing a captured image using a threshold value for enabling a marker to be identified, detecting a marker candidate from the binarized captured image, and excluding marker candidates detected by incorrect recognition from the detected marker candidates is known (see JP-A-2007-304732).
However, since the two-dimensional code illustrated in JP-A-2013-84031 has to be detected based on the area of the first color and the area of the second color in the captured image including the aforementioned two-dimensional code, there have been cases where the detection has failed due to external factors such as dirt in these areas, variation of the aforementioned area with time, or variation of image capturing conditions such as ambient light at the time of capturing the aforementioned captured image.
In the image detection method in JP-A-2007-304732, there have been cases where it has been difficult to have compatibility between improvement in the detection accuracy of the marker and shortening of the detection time of the marker without detecting the aforementioned marker by a detection method appropriate for each of plural markers included in a captured image.
In this image detection method, there are cases where a threshold value for binarizing the aforementioned captured image in which can detect all markers are included in the captured image cannot be selected as a threshold value appropriate for detecting the marker, and thus there are cases where not all the markers included in the aforementioned captured image can be detected. As a result, in the aforementioned image detection method, there are cases where it is difficult to improve the accuracy of detecting the marker from the aforementioned captured image.